AI Paper Atlas

Research map / 3D vision / 3D generation

SG2Loc: Sequential Visual Localization on 3D Scene Graphs

Nicole Damblon, Olga Vysotska, Federico Tombari, Marc Pollefeys, Daniel Barath

ICML 2026 · 3D vision › 3D generation · approach: Generative indoor scene synthesis · 0 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add depth score to semantic score (ablation)SG2Loc · 3RScan first 4 scenes, 5 framesR@25cm,10deg: 0.35 -> 0.54Improved
add RGB (SSIM) score on top of semantic+depth (ablation)SG2Loc · 3RScan first 4 scenes, 5 framesR@25cm,10deg: 0.54 -> 0.61Improved
add PnP refinement with RoMav2 matching (ablation)SG2Loc · 3RScan first 4 scenes, 5 framesR@10cm,5deg: 0.35 -> 0.50Improved
use proposed particle initialization instead of uniform 3D grid sampling (ablation)SG2Loc · 3RScan first 4 scenes, 5 framesR@25cm,10deg: 0.30 -> 0.35Improved
use adaptive KLD resampling instead of fixed particle count (ablation)SG2Loc · 3RScan first 4 scenes, 5 framesR@25cm,10deg: 0.30 -> 0.35Improved
raycast at max resolution (stride 1) in final pass (ablation)SG2Loc · 3RScan first 4 scenes, 5 framesR@25cm,10deg: 0.35 -> 0.33No clear effect
use longer query sequences for scene retrieval (1 to 50 frames)SceneGraphLoc extension · 3RScan 50 scenesRt@1: 0.69 -> 0.81Improved
use keyframing with per-keyframe evaluation instead of HLocSG2Loc · 3RScan keyframingmedian position error (m): 0.63 -> 0.24Improved
longer sequences (5 to 25 frames)SG2Loc · 3RScanjoint recall R@0.25m,2deg: 0.34 -> 0.44Improved
use scene graph map instead of image database (HLoc)SG2Loc · 3RScanstorage per scene (MB): 57.6 -> 9.8Improved
use SG2Loc vs HLoc on short sequencesSG2Loc · ScanNet 5 framesjoint recall R@0.25m,2deg: 0.53 -> 0.49Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy